ClawHub
Back to skill

Security audit

cosprompt

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed prompt-generation helper that searches character or topic context and formats image-generation prompts, with no evidence of hidden persistence, data exfiltration, or destructive behavior.

Before installing, note that the skill may use web browsing to research a requested character or topic and its tool permissions are broader than strictly necessary for plain prompt writing. Use it when you want structured image prompts, and review generated prompts for copyright, age-appropriateness, and language fit before reuse.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger metadata is very broad ('用户需要生成 AI 绘图提示词、角色海报 prompt、Cosplay 海报提示词') and does not define clear activation boundaries or exclusions. In an agent ecosystem, this can cause the skill to activate in loosely related conversations and unnecessarily gain access to tools like browser_use, Write, and Edit, increasing the chance of unintended tool use or prompt-scope hijacking.

Natural-Language Policy Violations

Medium
Confidence
86% confidence
Finding
The skill description and body strongly assume Chinese-language interaction/output, but there is no documented user opt-in or fallback behavior for other languages. This can override user language preference, degrade transparency, and in multi-skill settings may cause misaligned outputs or confuse downstream processing that expects the user's original language.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.